Development and validation of a risk model for identification of non-neutropenic, critically ill adult patients at high risk of invasive Candida infection: the Fungal Infection Risk Evaluation (FIRE) Study

Invasive candidiasis

Invasive candidiasis is an important fungal disease caused by Candida albicans and, increasingly, non-albicans Candida pathogens. Invasive Candida infections originate most frequently from endogenous human reservoirs and are triggered by impaired host defences. Signs and symptoms of invasive candidiasis are non-specific; candidaemia is the most diagnosed manifestation, with disseminated candidiasis affecting single or multiple organs. Diagnosis poses many challenges, and conventional culture techniques are frequently supplemented by non-culture-based assays. The attributable mortality from candidaemia and disseminated infections is ~30%. Fluconazole resistance is a concern for Nakaseomyces glabratus, Candida parapsilosis, and Candida auris and less so in Candida tropicalis infection; acquired echinocandin resistance remains uncommon. The epidemiology of invasive candidiasis varies in different geographical areas and within various patient populations. Risk factors include intensive care unit stay, central venous catheter use, broad-spectrum antibiotics use, abdominal surgery and immune suppression. Early antifungal treatment and central venous catheter removal form the cornerstones to decrease mortality. The landscape of novel therapeutics is growing; however, the application of new drugs requires careful selection of eligible patients as the spectrum of activity is limited to a few fungal species. Unanswered questions and knowledge gaps define future research priorities and a personalized approach to diagnosis and treatment of invasive candidiasis is of paramount importance.

Variation in Severity-Adjusted Resource use and Outcome for Neurosurgical Emergencies in the Intensive Care Unit

BACKGROUND

The correlation between the standardized resource use ratio (SRUR) and standardized hospital mortality ratio (SMR) for neurosurgical emergencies is not known. We studied SRUR and SMR and the factors affecting these in patients with traumatic brain injury (TBI), nontraumatic intracerebral hemorrhage (ICH), and subarachnoid hemorrhage (SAH).

METHODS

We extracted data of patients treated in six university hospitals in three countries (2015-2017). Resource use was measured as SRUR based on purchasing power parity-adjusted direct costs and either intensive care unit (ICU) length of stay (costSRURlength of stay) or daily Therapeutic Intervention Scoring System scores (costSRURTherapeutic Intervention Scoring System). Five a priori defined variables reflecting differences in structure and organization between the ICUs were used as explanatory variables in bivariable models, separately for the included neurosurgical diseases.

RESULTS

Out of 28,363 emergency patients treated in six ICUs, 6,162 patients (22%) were admitted with a neurosurgical emergency (41% nontraumatic ICH, 23% SAH, 13% multitrauma TBI, and 23% isolated TBI). The mean costs for neurosurgical admissions were higher than for nonneurosurgical admissions, and the neurosurgical admissions corresponded to 23.6-26.0% of all direct costs related to ICU emergency admissions. A higher physician-to-bed ratio was associated with lower SMRs in the nonneurosurgical admissions but not in the neurosurgical admissions. In patients with nontraumatic ICH, lower costSRURs were associated with higher SMRs. In the bivariable models, independent organization of an ICU was associated with lower costSRURs in patients with nontraumatic ICH and isolated/multitrauma TBI but with higher SMRs in patients with nontraumatic ICH. A higher physician-to-bed ratio was associated with higher costSRURs for patients with SAH. Larger units had higher SMRs for patients with nontraumatic ICH and isolated TBI. None of the ICU-related factors were associated with costSRURs in nonneurosurgical emergency admissions.

CONCLUSIONS

Neurosurgical emergencies constitute a major proportion of all emergency ICU admissions. A lower SRUR was associated with higher SMR in patients with nontraumatic ICH but not for the other diagnoses. Different organizational and structural factors seemed to affect resource use for the neurosurgical patients compared with nonneurosurgical patients. This emphasizes the importance of case-mix adjustment when benchmarking resource use and outcomes.

Can Beta-D-Glucan testing as part of the diagnostic pathway for Invasive Fungal Infection reduce anti-fungal treatment costs?

We performed a cost comparison of the current diagnostic and treatment pathway for invasive fungal infection (IFI) versus a proposed pathway that incorporates Beta-D-Glucan (BDG) testing from the NHS perspective. A fungal pathogen was identified in 58/107 (54.2%) patients treated with systemic anti-fungals in the Critical Care Department. Mean therapy duration was 23 days (standard deviation [SD] = 22 days), and cost was £5590 (SD = £7410) per patient. Implementation of BDG tests in the diagnostic and treatment pathway of patients with suspected IFI could result in a mean saving of £1643 per patient should a result be returned within two days.

Development and Validation of a Risk Score for Predicting Invasive Candidiasis in Intensive Care Unit Patients by Incorporating Clinical Risk Factors and Lymphocyte Subtyping

Objective

To develop and validate a rapid invasive candidiasis (IC)-predictive risk score in intensive care unit (ICU) patients by incorporating clinical risk factors and parameters of lymphocyte subtyping.

Methods

A prospective cohort study of 1054 consecutive patients admitted to ICU was performed. We assessed the clinical characteristics and parameters of lymphocyte subtyping at the onset of clinical signs of infection and their potential influence on IC diagnosis. A risk score for early diagnosis of IC was developed and validated based on a logistic regression model.

Results

Sixty-nine patients (6.5%) had IC. Patients in the cohort (N=1054) were randomly divided into a development (n=703) or validation (n=351) cohorts. Multivariate logistic regression identified that CD8+ T-cell count ≤143 cells/mm³, receipt of high-dose corticosteroids (dose ≥50 mg prednisolone equivalent), receipt of carbapenem/tigecycline, APACHE II score≥15, (1,3)-β-D-glucan (BDG) positivity and emergency gastrointestinal/hepatobiliary (GIT/HPB) surgery were significantly related with IC. IC risk score was calculated using the following formula: CD8+ T-cell count ≤143 cells/mm³ + receipt of high-dose corticosteroids + receipt of carbapenem/tigecycline + APACHE II score ≥15 + BDG positivity + emergency GIT/HPB surgery ×2. The risk scoring system had good discrimination and calibration with area under the receiver operating characteristic (AUROC) curve of 0.820 and 0.807, and a non-significant Hosmer-Lemeshow test P=0.356 and P=0.531 in the development and validation cohorts, respectively. We categorized patients into three groups according to risk score: low risk (0-2 points), moderate risk (3-4 points) and high risk (5-7 points). IC risk was highly and positively associated with risk score (Pearson contingency coefficient=0.852, P for trend=0.007). Candida score had a moderate predicting efficacy for early IC diagnosis. The AUROC of the risk score was significantly larger than that of Candida score (0.820 versus 0.711, Z=2.013, P=0.044).

Conclusions

The predictive scoring system, which used both clinical factors and CD8+ T cell count, served as a clinically useful predictive model for rapid IC diagnosis in this cohort of ICU patients.

Clinical Trial Registration

chictr.org.cn, identifier ChiCTR-ROC-17010750.

Risk factors for invasive candida infection in critically ill patients - a systematic review and meta-analysis

Background

Current guidelines recommend empirical antifungal therapy in patients with sepsis with high risk of invasive Candida infection. However, many different risk factors have been derived from multiple studies. These risk factors lack specificity, and broad application would render most ICU patients eligible for empirical antifungal therapy.

Research Question

What risk factors for invasive Candida infection can be identified by a systematic review and meta-analysis?

Study Design and Methods

We searched PubMed, Web of Science, ScienceDirect, Biomed Central, and Cochrane and extracted the raw and adjusted OR for each risk factor associated with invasive Candida infection. We calculated pooled ORs for risk factors present in more than one study.

Results

We included 34 studies in our meta-analysis resulting in the assessment of 29 possible risk factors. Risk factors for invasive Candida infection included demographic factors, comorbid conditions, and medical interventions. Although demographic factors do not play a role for the development of invasive Candida infection, comorbid conditions (eg, HIV, Candida colonization) and medical interventions have a significant impact. The risk factors associated with the highest risk for invasive Candida infection were broad-spectrum antibiotics (OR, 5.6; 95% CI, 3.6-8.8), blood transfusion (OR, 4.9; 95% CI, 1.5-16.3), Candida colonization (OR, 4.7; 95% CI, 1.6-14.3), central venous catheter (OR, 4.7; 95% CI, 2.7-8.1), and total parenteral nutrition (OR, 4.6; 95% CI, 3.3-6.3). However, dependence between the various risk factors is probably high.

Interpretation

Our systematic review and meta-analysis identified patient- and treatment-related factors that were associated with the risk for the development of invasive Candida infection in the ICU. Most of the factors identified were either related to medical interventions during intensive care or to comorbid conditions.

Economic evaluations of big data analytics for clinical decision-making: a scoping review

OBJECTIVE

Much has been invested in big data analytics to improve health and reduce costs. However, it is unknown whether these investments have achieved the desired goals. We performed a scoping review to determine the health and economic impact of big data analytics for clinical decision-making.

MATERIALS AND METHODS

We searched Medline, Embase, Web of Science and the National Health Services Economic Evaluations Database for relevant articles. We included peer-reviewed papers that report the health economic impact of analytics that assist clinical decision-making. We extracted the economic methods and estimated impact and also assessed the quality of the methods used. In addition, we estimated how many studies assessed "big data analytics" based on a broad definition of this term.

RESULTS

The search yielded 12 133 papers but only 71 studies fulfilled all eligibility criteria. Only a few papers were full economic evaluations; many were performed during development. Papers frequently reported savings for healthcare payers but only 20% also included costs of analytics. Twenty studies examined "big data analytics" and only 7 reported both cost-savings and better outcomes.

DISCUSSION

The promised potential of big data is not yet reflected in the literature, partly since only a few full and properly performed economic evaluations have been published. This and the lack of a clear definition of "big data" limit policy makers and healthcare professionals from determining which big data initiatives are worth implementing.

Budget Impact of Microbial Cell-Free DNA Testing Using the Karius® Test as an Alternative to Invasive Procedures in Immunocompromised Patients with Suspected Invasive Fungal Infections

BACKGROUND

Invasive fungal infection is a major source of morbidity and mortality. The usage of microbial cell-free DNA for the detection and identification of invasive fungal infection has been considered as a potential alternative to invasive procedures allowing for rapid results.

OBJECTIVE

This analysis aimed to assess the budget implications of using the Karius^® Test in patients suspected of invasive fungal infection in an average state in the USA from a healthcare payer perspective.

METHODS

The analysis used a decision tree to capture key stages of the patient pathway, from suspected invasive fungal infection to either receiving treatment for invasive fungal infection or being confirmed as having no invasive fungal infection. The analysis used published costs and resource use from a targeted review of the literature. Because of the paucity of published evidence on the reduction of diagnostic tests displaced by the Karius Test, the analysis used a 50% reduction in the use of bronchoscopy and/or bronchoalveolar lavage. The impact of this reduction was tested in a scenario analysis.

RESULTS

The results of the analysis show that the introduction of the Karius Test is associated with a cost saving of US$2277 per patient; when multiplied by the estimated number of cases per year, the cost saving is US$17,039,666. The scenario analysis showed that the Karius Test only had an incremental cost of US$87 per patient when there was no reduction in bronchoscopy and bronchoalveolar lavage.

CONCLUSIONS

The Karius Test may offer a valuable and timely option for the diagnosis of invasive fungal infection through its non-invasive approach and subsequent cost savings.

Invasive candidiasis in critical care: challenges and future directions

Invasive candidiasis is the most common critical care-associated fungal infection with a crude mortality of ~ 40-55%. Important factors contributing to risk of invasive candidiasis in ICU include use of broad-spectrum antimicrobials, immunosuppressive drugs, and total parenteral nutrition alongside iatrogenic interventions which breach natural barriers to infection [vascular catheters, renal replacement therapy, extracorporeal membrane oxygenation (ECMO), surgery]. This review discusses three key challenges in this field. The first is the shift in Candida epidemiology across the globe to more resistant non-albicans species, in particular, the emergence of multi-resistant Candida glabrata and Candida auris, which pose significant treatment and infection control challenges in critical care. The second challenge lies in the timely and appropriate initiation and discontinuation of antifungal therapy. Early antifungal strategies (prophylaxis, empirical and pre-emptive) using tools such as the Candida colonisation index, clinical prediction rules and fungal non-culture-based tests have been developed: we review the evidence on implementation of these tools in critical care to aid clinical decision-making around the prescribing and cessation of antifungal therapy. The third challenge is selection of the most appropriate antifungal to use in critical care patients. While guidelines exist to aid choice, this heterogenous and complex patient group require a more tailored approach, particularly in cases of acute kidney injury, liver impairment and for patients supported by extracorporeal membrane oxygenation. We highlight key research priorities to overcome these challenges in the future.

Admission patterns and survival from status epilepticus in critical care in the UK: an analysis of the Intensive Care National Audit and Research Centre Case Mix Programme database

BACKGROUND AND PURPOSE

Factors influencing the outcome after the critical care unit (CCU) for patients with status epilepticus (SE) are poorly understood. Survival for these patients was examined to establish (i) whether the risk of mortality has changed over time and (ii) whether admission to different unit types affects mortality risk over and above other risk factors.

METHODS

The Intensive Care National Audit and Research Centre database and the Case Mix Programme database (January 2001 to December 2016) were analysed. Units were defined as neuro-CCU (NCCU), general CCU with 24-h neurological support (GCCU-N) or general CCU with limited neurological support (GCCU-L).

RESULTS

There were 35 595 CCU cases of SE with a 3-fold increase over time (4739 in 2001-2004 to 14 166 in 2013-2016). More recent admissions were older and were more often unsedated on admission. Mortality declined for all units although this was more marked for NCCUs (8.1% in 2001-2004 to 4.4% in 2013-2016 compared to 5.1% and 4.1% for GCCU-L). Acute hospital mortality was two to three times higher than CCU mortality although this has also declined with time. GCCU-L appeared to have lower mortality than NCCUs (odds ratio 0.84, 95% confidence interval 0.72, 0.98) but after post hoc adjustment for case mix there were no differences. Older age and markers of seriousness of morbidity were all associated with increased mortality risk.

CONCLUSIONS

The number of patients admitted to a CCU for SE is rising but critical care and acute hospital mortality is decreasing. Patients treated in an NCCU have higher mortality but this is explicable by more severe underlying disease.

Recent advances and novel approaches in laboratory-based diagnostic mycology

The field of diagnostic mycology represents much more than culture and microscopy and is rapidly embracing novel techniques and strategies to help overcome the limitations of conventional approaches. Commercial molecular assays increase the applicability of PCR testing and may identify markers of antifungal resistance, which are of great clinical concern. Lateral flow assays simplify testing and turn-around time, with potential for point of care testing, while proximity ligation assays embrace the sensitivity of molecular testing with the specificity of antibody detection. The first evidence of patient risk stratification is being described and together with the era of next generation sequencing represents an exciting time in mycology.

The epidemiology and management of candidemia in Northern Ireland during 2002-2011, including a 12-month enhanced case review

In Northern Ireland there are concerns about candidaemia, with rates higher than those reported in England and Wales. Our aim was to explore the epidemiology of candidaemia during a 10 year period and the clinical management upon suspicion of cases during a one year enhanced investigation in Northern Ireland.Candidaemia reports to the Public Health Agency were validated during 2002-2011 and used to examine incidence and antifungal sensitivity trends (during 2007-2011). A clinical proforma was used to collate information for all patients with candidaemia in 2011.The majority (96%) of isolates were captured through voluntary laboratory reporting. There was a year-on-year increase in candidaemia from 2002-2011, from 80 to 131 episodes (incidence rate ratio 1.09 95% CI 1.05-1.13). Rates were highest in males under 1 year and over 75 years. 83/98 (85%) of case notes were available from candidaemia patients during 2011. The most prevalent risk factors were patients on total parenteral nutrition (26 people, 31.3%), surgery in the two months prior to the candidaemia (25 people, 30.1%), significant steroid use in the previous 3 months (24 people, 28.9%) and active neoplastic disease (23 people, 27.7%),This study confirmed an increase in candidaemia rates over time, with the observed incidence in 2011 higher than England and Wales. We identified areas for improvement around the clinical management of candidaemia. We recommend raising the awareness of guidelines for fundoscopy, echocardiography and central venous catheter removal.

Handling of Missing Outcome Data in Traumatic Brain Injury Research: A Systematic Review

Traumatic brain injury (TBI) research commonly measures long-term functional outcome, but studies often suffer from missing data as patients are lost to follow-up. This review assesses the extent and handling of missing outcome data in the TBI literature and provides a practical guide for future research. Relevant electronic databases were searched from January 1, 2012 to October 27, 2017 for TBI studies that used the Glasgow Outcome Scale or Glasgow Outcome Scale-Extended (GOS/GOSE) as an outcome measure. Studies were screened and data extracted in line with Cochrane guidance.

A total of 195 studies, 21 interventional, 174 observational, with 104,688 patients were included. Using the reported follow-up rates in a mixed model, on average 91% of patients were predicted to return to follow-up at 6 months post-injury, 84% at 1 year, and 69% at 2 years. However, 36% of studies provided insufficient information to determine the number of subjects at each time-point. Of 139 studies that did report missing outcome data, only 50% attempted to identify why data were missing, with just 4 reporting their assumption on the “missingness mechanism.” The handling of missing data was heterogeneous, with the most common method being its exclusion from analysis. These results confirm substantial variability in the standard of reporting and handling of missing outcome data in TBI research. We conclude that practical guidance is needed to facilitate meaningful and accurate study interpretation, and therefore propose a framework for the handling of missing outcome data in future TBI research.

Invasive pulmonary aspergillosis in cirrhotic patients: analysis of a 10-year clinical experience

BACKGROUND

Cirrhosis is not recognised as one of the main risk factors of invasive pulmonary aspergillosis (IPA), although its prevalence is increasing. The aim of our study was to identify factors for IPA in such patients with a positive Aspergillus sp. culture in respiratory samples and to evaluate its impact on outcome.

METHODS

We conducted a monocentric retrospective study between January 2005 and December 2015. All cirrhotic patients hospitalised in our liver ICU with a positive Aspergillus sp. respiratory sample were included. These patients were case-matched with cirrhotic patients without positive Aspergillus respiratory sample. Finally, the patients were classified as having putative aspergillosis or colonisation according to the criteria described previously.

RESULTS

In total, 986 cirrhotic patients were admitted to ICU during the study period. Among these, sixty patients had a positive Aspergillus sp. respiratory sample. Chronic obstructive pulmonary disease (COPD) comorbidity and organ supports were significantly associated with Aspergillus colonisation. Seventeen patients (28%) were diagnosed as proven or putative IPA and 43 were considered as colonised by Aspergillus sp. The median delay between ICU admission and an IPA diagnosis was 2 [2-24] days. Only COPD was predictive of the presence of IPA (OR 6.44; 95% CI 1.43-28.92; p = 0.0151) in patients with a positive Aspergillus sp. culture. The probability of in-hospital mortality was 71% in the IPA group versus 19% in the colonisation group (p = 0.0001).

CONCLUSION

Patients with cirrhosis can be at risk of IPA, especially with COPD. Antifungal agents should be given as soon as possible mainly in cirrhotic patients with COPD.

A Risk Prediction Model for Invasive Fungal Disease in Critically Ill Patients in the Intensive Care Unit

PURPOSE

Developing a risk prediction model for invasive fungal disease based on an analysis of the disease-related risk factors in critically ill patients in the intensive care unit (ICU) to diagnose the invasive fungal disease in the early stages and determine the time of initiating early antifungal treatment.

METHODS

Data were collected retrospectively from 141 critically ill adult patients with at least 4 days of general ICU stay at Sun Yat-sen Memorial Hospital, Sun Yat-sen University during the period from February 2015 to February 2016. Logistic regression was used to develop the risk prediction model. Discriminative power was evaluated by the area under the receiver operating characteristics (ROC) curve (AUC).

RESULTS

Sequential organ failure assessment (SOFA) score, antibiotic treatment period, and positive culture of Candida albicans other than normally sterile sites are the three predictors of invasive fungal disease in critically ill patients in the ICU. The model performs well with an ROC-AUC of .73.

CONCLUSION

The risk prediction model performs well to discriminate between critically ill patients with or without invasive fungal disease. Physicians could use this prediction model for early diagnosis of invasive fungal disease and determination of the time to start early antifungal treatment of critically ill patients in the ICU.

Identification of medically important Candida species by polymerase chain reaction-restriction fragment length polymorphism analysis of the rDNA ITS1 and ITS2 regions

AIM

We aimed to identify the distribution of species in candidal strains isolated from clinical samples and restriction fragment length polymorphism (RFLP) method based on Msp I and Bln I restrictive enzyme cuts of polymerase chain reaction (PCR) products after the amplification of ITS1 and ITS2 regions of rDNA genotypically.

MATERIALS AND METHODS

One hundred and fifty candidal strains isolated from various clinical samples were studies/ included. Phenotypic species assessment was performed using automated VITEK-2 system and kit used with the biochemical tests. Common genomic region amplification peculiar to candidal strains was carried out using ITS1 and ITS2 primer pairs. After the amplification, PCR products were cut with Msp I and Bln I restriction enzymes for species identification.

RESULTS

The majority of Candida isolates were isolated from urine (78.6%) while other isolates were composed of strains isolated from swab, wound, blood and other samples by 11.3%, 3.3%, 2% and 4.7%, respectively. The result of RFLP analysis carried out with Msp I and Bln I restriction enzymes showed that candidal strains were Candida albicans by 45.3%, Candida glabrata by 19.3%, Candida tropicalis by 14.6%, Candida parapsilosis by 5.3%, Candida krusei by 5.3%, Candida lusitaniae by 0.6% and other candidal strains by 9.3%.

CONCLUSION

When the ability to identify Candida to species level of phenotypic and PCR-RFLP methods was assessed, a great difference was found between these two methods. It may be argued that Msp I and Bln I restriction enzyme fragments can be used in the identification of medically important Candida species. Further studies are needed to develop this kind of restriction profile to be used in the identification of candidal strains.

Sepsis - thoughtful management for the non-expert

Sepsis is a common and often devastating medical emergency with a high mortality rate and, in many survivors, long-term morbidity. It is defined as the dysregulated host response to infection resulting in organ dysfunction, and its incidence is increasing as the population ages. However, it is a treatable and potentially reversible condition, especially if identified and treated promptly. A sound understanding of sepsis is crucial for optimal care. Although general guidelines are available for management, here we provide a foundation of understanding to encourage thoughtful, personalised management of sepsis during the acute phase. We provide an overview of the epidemiology, the new Sepsis-3 definitions, pathophysiology, clinical presentations, and investigation and management of sepsis for the non-expert.

Identifying Important Attributes for Prognostic Prediction in Traumatic Brain Injury Patients

Background: Generally, traumatic brain injury (TBI) patients do not have a stable condition, particularly after the first week of TBI. Hence, indicating the attributes in prognosis through a prediction model is of utmost importance since it helps caregivers with treatment-decision options, or prepares the relatives for the most-likely outcome. Objectives: This study attempted to determine and order the attributes in prognostic prediction in TBI patients, based on early clinical findings. A hybrid method was employed, which combines a decision tree (DT) and an artificial neural network (ANN) in order to improve the modeling process. Methods: The DT approach was applied as the initial analysis of the network architecture to increase accuracy in prediction. Afterwards, the ANN structure was mapped from the initial DT based on a part of the data. Subsequently, the designed network was trained and validated by the remaining data. 5-fold cross-validation method was applied to train the network. The area under the receiver operating characteristic (ROC) curve, sensitivity, specificity, and accuracy rate were utilized as performance measures. The important attributes were then determined from the trained network using two methods: change of mean squared error (MSE), and sensitivity analysis (SA). Results: The hybrid method offered better results compared to the DT method. The accuracy rate of 86.3 % vs. 82.2 %, sensitivity value of 55.1 % vs. 47.6 %, specificity value of 93.6 % vs. 91.1 %, and the area under the ROC curve of 0.705 vs. 0.695 were achieved for the hybrid method and DT, respectively. However, the attributes’ order by DT method was more consistent with the clinical literature. Conclusions: The combination of different modeling methods can enhance their performance. However, it may create some complexities in computations and interpretations. The outcome of the present study could deliver some useful hints in prognostic prediction on the basis of early clinical findings for TBI patients.

Risk factors for health care-associated infection in hospitalized adults: Systematic review and meta-analysis

BACKGROUND

Health care-associated infections (HAIs) are a public health problem that increase health care costs. This article aimed to systematically review the literature and meta-analyze studies investigating risk factors (RFs) independently associated with HAIs in hospitalized adults.

METHODS

Electronic databases (MEDLINE, Embase, and LILACS) were searched to identify studies from 2009-2016. Pooled risk ratios (RRs) or odds ratios (ORs) or mean differences (MDs) and 95% confidence intervals (CIs) were calculated and compared across the groups. This review followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses statement.

RESULTS

Of 867 studies, 65 met the criteria for review, and the data of 18 were summarized in the meta-analysis. The major RFs independently associated with HAIs were diabetes mellitus (RR, 1.76; 95% CI, 1.27-2.44), immunosuppression (RR, 1.24; 95% CI, 1.04-1.47), body temperature (MD, 0.62; 95% CI, 0.41-0.83), surgery time in minutes (MD, 34.53; 95% CI, 22.17-46.89), reoperation (RR, 7.94; 95% CI, 5.49-11.48), cephalosporin exposure (RR, 1.77; 95% CI, 1.30-2.42), days of exposure to central venous catheter (MD, 5.20; 95% CI, 4.91-5.48), intensive care unit (ICU) admission (RR, 3.76; 95% CI, 1.79-7.92), ICU stay in days (MD, 21.30; 95% CI, 19.81-22.79), and mechanical ventilation (OR, 12.95; 95% CI, 6.28-26.73).

CONCLUSIONS

Identifying RFs that contribute to develop HAIs may support the implementation of strategies for their prevention, therefore maximizing patient safety.

Enhancement of hepatitis virus immunoassay outcome predictions in imbalanced routine pathology data by data balancing and feature selection before the application of support vector machines

Background

Data mining techniques such as support vector machines (SVMs) have been successfully used to predict outcomes for complex problems, including for human health. Much health data is imbalanced, with many more controls than positive cases.

Methods

The impact of three balancing methods and one feature selection method is explored, to assess the ability of SVMs to classify imbalanced diagnostic pathology data associated with the laboratory diagnosis of hepatitis B (HBV) and hepatitis C (HCV) infections. Random forests (RFs) for predictor variable selection, and data reshaping to overcome a large imbalance of negative to positive test results in relation to HBV and HCV immunoassay results, are examined. The methodology is illustrated using data from ACT Pathology (Canberra, Australia), consisting of laboratory test records from 18,625 individuals who underwent hepatitis virus testing over the decade from 1997 to 2007.

Results

Overall, the prediction of HCV test results by immunoassay was more accurate than for HBV immunoassay results associated with identical routine pathology predictor variable data. HBV and HCV negative results were vastly in excess of positive results, so three approaches to handling the negative/positive data imbalance were compared. Generating datasets by the Synthetic Minority Oversampling Technique (SMOTE) resulted in significantly more accurate prediction than single downsizing or multiple downsizing (MDS) of the dataset. For downsized data sets, applying a RF for predictor variable selection had a small effect on the performance, which varied depending on the virus. For SMOTE, a RF had a negative effect on performance. An analysis of variance of the performance across settings supports these findings. Finally, age and assay results for alanine aminotransferase (ALT), sodium for HBV and urea for HCV were found to have a significant impact upon laboratory diagnosis of HBV or HCV infection using an optimised SVM model.

Conclusions

Laboratories looking to include machine learning via SVM as part of their decision support need to be aware that the balancing method, predictor variable selection and the virus type interact to affect the laboratory diagnosis of hepatitis virus infection with routine pathology laboratory variables in different ways depending on which combination is being studied. This awareness should lead to careful use of existing machine learning methods, thus improving the quality of laboratory diagnosis.

A prospective study of fungal biomarkers to improve management of invasive fungal diseases in a mixed specialty critical care unit

PURPOSE

The diagnosis of invasive fungal diseases (IFD) in critical care patients (CrCP) is difficult. The study investigated the performance of a set of biomarkers for diagnosis of IFD in a mixed specialty critical care unit (CrCU).

METHODS

A prospective observational study in patients receiving critical care for ≥7days was performed. Serum samples were tested for the presence of: (1-3) - β-d-glucan (BDG), galactomannan (GM), and Aspergillus fumigatus DNA. GM antigen detection was also performed on bronchoalveolar lavage (BAL) samples. The patients were classified using published definitions for IFD and a diagnostic algorithm for invasive pulmonary aspergillosis. Performance parameters of the assays were determined.

RESULTS

In patients with proven and probable IFD, the sensitivity, specificity, PPV and NPV of a single positive BDG were 63%, 83%, 65% and 83% respectively. Specificity increased to 86% with 2 consecutive positive results. The mean BDG value of patients with proven and probable IFD was significantly higher compared to those with fungal colonization and no IFD (p value<0.0001).

CONCLUSION

New diagnostic criteria which incorporate these biomarkers, in particular BDG, and host factors unique to critical care patients should enhance diagnosis of IFD and positively impact antifungal stewardship programs.

Predicting invasive fungal disease due to Candida species in non-neutropenic, critically ill, adult patients in United Kingdom critical care units

Background

Given the predominance of invasive fungal disease (IFD) amongst the non-immunocompromised adult critically ill population, the potential benefit of antifungal prophylaxis and the lack of generalisable tools to identify high risk patients, the aim of the current study was to describe the epidemiology of IFD in UK critical care units, and to develop and validate a clinical risk prediction tool to identify non-neutropenic, critically ill adult patients at high risk of IFD who would benefit from antifungal prophylaxis.

Methods

Data on risk factors for, and outcomes from, IFD were collected for consecutive admissions to adult, general critical care units in the UK participating in the Fungal Infection Risk Evaluation (FIRE) Study. Three risk prediction models were developed to model the risk of subsequent Candida IFD based on information available at three time points: admission to the critical care unit, at the end of 24 h and at the end of calendar day 3 of the critical care unit stay. The final model at each time point was evaluated in the three external validation samples.

Results

Between July 2009 and April 2011, 60,778 admissions from 96 critical care units were recruited. In total, 359 admissions (0.6 %) were admitted with, or developed, Candida IFD (66 % Candida albicans). At the rate of candidaemia of 3.3 per 1000 admissions, blood was the most common Candida IFD infection site. Of the initial 46 potential variables, the final admission model and the 24-h model both contained seven variables while the end of calendar day 3 model contained five variables. The end of calendar day 3 model performed the best with a c index of 0.709 in the full validation sample.

Conclusions

Incidence of Candida IFD in UK critical care units in this study was consistent with reports from other European epidemiological studies, but lower than that suggested by previous hospital-wide surveillance in the UK during the 1990s. Risk modeling using classical statistical methods produced relatively simple risk models, and associated clinical decision rules, that provided acceptable discrimination for identifying patients at ‘high risk’ of Candida IFD.

Trial registration

The FIRE Study was reviewed and approved by the Bolton NHS Research Ethics Committee (reference: 08/H1009/85), the Scotland A Research Ethics Committee (reference: 09/MRE00/76) and the National Information Governance Board (approval number: PIAG 2-10(f)/2005).

Electronic supplementary material

The online version of this article (doi:10.1186/s12879-016-1803-9) contains supplementary material, which is available to authorized users.

An evaluation of the clinical and cost-effectiveness of alternative care locations for critically ill adult patients with acute traumatic brain injury

BACKGROUND

For critically ill adult patients with acute traumatic brain injury (TBI), we assessed the clinical and cost-effectiveness of: (a) Management in dedicated neurocritical care units versus combined neuro/general critical care units within neuroscience centres. (b) 'Early' transfer to a neuroscience centre versus 'no or late' transfer for those who present at a non-neuroscience centre.

METHODS

The Risk Adjustment In Neurocritical care (RAIN) Study included prospective admissions following acute TBI to 67 UK adult critical care units during 2009-11. Data were collected on baseline case-mix, mortality, resource use, and at six months, Glasgow Outcome Scale Extended (GOSE), and quality of life (QOL) (EuroQol 5D-3L). We report incremental effectiveness, costs and cost per Quality-Adjusted Life Year (QALY) of the alternative care locations, adjusting for baseline differences with validated risk prediction models. We tested the robustness of results in sensitivity analyses.

FINDINGS

Dedicated neurocritical care unit patients (N = 1324) had similar six-month mortality, higher QOL (mean gain 0.048, 95% CI -0.002 to 0.099) and increased average costs compared with those managed in combined neuro/general units (N = 1341), with a lifetime cost per QALY gained of £14,000. 'Early' transfer to a neuroscience centre (N = 584) was associated with lower mortality (odds ratio 0.52, 0.34-0.80), higher QOL for survivors (mean gain 0.13, 0.032-0.225), but positive incremental costs (£15,001, £11,123 to £18,880) compared with 'late or no transfer' (N = 263). The lifetime cost per QALY gained for 'early' transfer was £11,000.

CONCLUSIONS

For critically ill adult patients with acute TBI, within neuroscience centres management in dedicated neurocritical care units versus combined neuro/general units led to improved QoL and higher costs, on average, but these differences were not statistically significant. This study finds that 'early' transfer to a neuroscience centre is associated with reduced mortality, improvement in QOL and is cost-effective.

Clinical Results and Outcome Improvement Over Time in Traumatic Brain Injury

Prognostic models for traumatic brain injury (TBI) are important tools both in clinical practice and research if properly validated, preferably by external validation. Prognostic models also offer the possibility of monitoring performance by comparing predicted outcomes with observed outcomes. In this study, we applied the prognostic models developed by the International Mission on Prognosis and Analysis of Clinical Trials in TBI (IMPACT) in an Italian multi-center database (Neurolink) with two aims: to compare observed with predicted outcomes and to check for a possible improvement of clinical outcome over the 11 years of patient inclusion in Neurolink. We applied the IMPACT models to patients included in Neurolink between 1997 and 2007. Performance of the models was assessed by determining calibration (with calibration plots) and discrimination (by the area under the receiver operating characteristic curve [AUC]). Logistic regression analysis was used to analyze a possible trend in outcomes over time, adjusted for predicted outcomes. A total of 1401 patients were studied. Patients had a median age of 40 years and 51% had a Glasgow Coma Scale motor score of 5 or 6. The models showed good discrimination, with AUCs of 0.86 (according to the Core Model) and 0.88 (Extended Model), and adequate calibration, with the overall observed risk of unfavorable outcome and mortality being less than predicted. Outcomes significantly improved over time. This study shows that the IMPACT models performed reasonably well in the Neurolink data and can be used for monitoring performance. After adjustment for predicted outcomes with the prognostic models, we observed a substantial improvement of patient outcomes over time in the three Neurolink centers.

The Monitoring and Management of Severe Traumatic Brain Injury in the United Kingdom: Is there a Consensus?: A National Survey

BACKGROUND

To survey the current practice of monitoring and management of severe traumatic brain injury (TBI) patients in the critical care units across the United Kingdom.

METHODS

A structured telephone interview was conducted with senior medical or nursing staff of all the adult neurocritical care units. Thirty-one neurocritical care units that managed adult patients with severe TBI were identified from the Risk Adjustment in Neurocritical Care (RAIN) study and the Society of British Neurological Surgeons.

RESULTS

Intracranial pressure (ICP) monitoring was used in all the 31 institutions. Cerebral perfusion pressure was used in 30 of the 31 units and a Cerebral perfusion pressure target of 60 to 70 mm Hg was the most widely used target (25 of 31 units). Transcranial Doppler was used in 12 units (39%); brain tissue oxygen (PbtO(2)) was used in 8 (26%); cerebral microdialysis was used in 4 (13%); jugular bulb oximetry in 1 unit; and near-infrared spectrometry was not used in any unit. Continuous capnometry was used in 28 (91%) units for mechanically ventilated patients. Mannitol was the most commonly used agent for osmotherapy to treat intracranial hypertension.

CONCLUSIONS

We identified that there was no clear consensus and considerable variation in practice in the management of TBI patients in UK neurocritical care units. A protocol-based management has been shown to improve outcome in sepsis patients. Given the magnitude of the problem, we conclude that there is an urgent need for international consensus guidelines for management of TBI patients in critical care units.

Predicting the outcome for individual patients with traumatic brain injury: a case-based review

BACKGROUND

Traumatic brain injuries result in significant morbidity and mortality. Accurate prediction of prognosis is desirable to inform treatment decisions and counsel family members. Objective To review the currently available prognostic tools for use in traumatic brain injury (TBI), to analyse their value in individual patient management and to appraise ongoing research on prognostic modelling.

METHODS AND RESULTS

We present two patients who sustained a TBI in 2011-2012 and evaluate whether prognostic models could accurately predict their outcome. The methodology and validity of current prognostic models are analysed and current research that might contribute to improved individual patient prognostication is evaluated.

CONCLUSION

Predicting prognosis in the acute phase after TBI is complex and existing prognostic models are not suitable for use at the individual patient level. Data derived from these models should only be used as an adjunct to clinical judgement and should not be used to set limits for acute care interventions. Information from neuroimaging, physiological monitoring and analysis of biomarkers or genetic polymorphisms may be used in the future to improve accuracy of individual patient prognostication. Clinicians should consider offering full supportive treatment to patients in the early phase after injury whilst the outcome is unclear.

Early invasive fungal infections and colonization in patients with cirrhosis admitted to the intensive care unit

Bacterial infections in cirrhosis are common and associated with increased mortality, but little is known about fungal infections. The aim of this study, a sub-analysis of the Fungal Infection Risk Evaluation study, was to assess the incidence and implications of early invasive fungal disease (IFD) in patients with cirrhosis admitted to intensive care units (ICU). Clinical and laboratory parameters collected in the first 3 days of ICU stay for 782 patients with cirrhosis and/or portal hypertension were analysed and compared with those of 273 patients with very severe cardiovascular disease (CVD). The CVD patients had more co-morbidities and higher APACHE II scores. The overall incidence of IFD was similar in the two groups, but the incidence of IFD in ICU was higher in liver patients (1% versus 0.4%; p 0.025) as was fungal colonization (23.8% versus 13.9%; p 0.001). The ICU and in-hospital mortality, and length of stay were similar in the two groups. A higher proportion of liver patients received antifungal therapy (19.2% versus 7%; p <0.0005). There was no difference in mortality between colonized patients who received antifungal therapy and colonized patients who did not. The incidence of IFD in patients with cirrhosis in ICU is higher compared with another high-risk group, although it is still very low. This risk might be higher in patients with advanced liver disease admitted with acute-on-chronic liver failure, and this should be investigated further. Our data do not support prophylactic use of antifungal therapy in cirrhosis.

Invasive candidiasis in non neutropenic critically ill - need for region-specific management guidelines

Use of antifungal agents has increased over past few decades. A number of risk factors such as immunosuppression, broad spectrum antibiotics, dialysis, pancreatitis, surgery, etc., have been linked with the increased risk of invasive candidiasis. Though there are various guidelines available for the use of antifungal therapy, local/regional epidemiology plays an important role in determining the appropriate choice of agent in situations where the offending organism is not known (i.e. empirical, prophylactic or preemptive therapy). Developing countries like India need to generate their own epidemiological data to facilitate appropriate use of antifungal therapy. In this article, the authors have highlighted the need for region-specific policies/guidelines for treatment of invasive candidiasis. Currently available Indian literature on candidemia epidemiology has also been summarized here.

Invasive candidiasis in critical care setting, updated recommendations from “Invasive Fungal Infections-Clinical Forum”, Iran

Invasive candidiasis (IC) bears a high risk of morbidity and mortality in the intensive care units (ICU). With the current advances in critical care and the use of wide-spectrum antibiotics, invasive fungal infections (IFIs) and IC in particular, have turned into a growing concern in the ICU. Further to blood cultures, some auxiliary laboratory tests and biomarkers are developed to enable an earlier detection of infection, however these test are neither consistently available nor validated in our setting. On the other hand, patients’ clinical status and local epidemiology data may justify the empiric antifungal approach using the proper antifungal option. The clinical approach to the management of IC in febrile, non-neutropenic critically ill patients has been defined in available international guidelines; nevertheless such recommendations need to be customized when applied to our local practice. Over the past three years, Iranian experts from intensive care and infectious diseases disciplines have tried to draw a consensus on the management of IFI with a particular focus on IC in the ICU. The established IFI-clinical forum (IFI-CF), comprising the scientific leaders in the field, has recently come up with and updated recommendation on the same (June 2014). The purpose of this review is to put together literature insights and Iranian experts’ opinion at the IFI-CF, to propose an updated practical overview on recommended approaches for the management of IC in the ICU.

Risk prediction for invasive candidiasis

Over past few years, treatment of invasive candidiasis (IC) has evolved from targeted therapy to prophylaxis, pre-emptive and empirical therapy. Numerous predisposing factors for IC have been grouped together in various combinations to design risk prediction models. These models in general have shown good negative predictive value, but poor positive predictive value. They are useful in selecting the population which is less likely to benefit from empirical antifungal therapy and thus prevent overuse of antifungal agents. Current article deals with various risk prediction models for IC and their external validation studies.

Patterns of surgical care and complications in elderly adults

OBJECTIVES

To determine whether procedures, hospitals visited, and complications would differ according to decade in elderly adults and from those of younger adults.

DESIGN

Retrospective cohort study.

SETTING

The Anesthesia Quality Institute National Anesthesia Clinical Outcomes Registry (NACOR) is the largest database of anesthesia cases from academic and community hospitals and includes all insurance and facility types across the United States.

PARTICIPANTS

Eight million six hundred thirty-two thousand nine hundred seventy-nine cases from January 2010 to March 2013 were acquired. After exclusion of individuals younger than 18, nonapplicable locations, and brain death, 2,851,114 remained and were placed into age categories (18-64, 65-69, 70-79, 80-89, ≥ 90).

MEASUREMENTS

Participant, surgical, anesthetic, and hospital descriptors and short-term outcomes (major complications, mortality at <48 hours).

RESULTS

The largest number of older adults had surgery in medium-sized community hospitals. The oldest age group (≥ 90) underwent the smallest range of procedures; hip fracture, hip replacement, and cataract procedures accounted for more than 35% of all surgeries. Younger old adults underwent these procedures plus a significant proportion of spinal fusion, cholecystectomy, and knee surgery. Older adults had greater mortality and more complications than younger adults. Participants undergoing exploratory laparotomy had the greatest likelihood of death in any age category except 90 and older, in which small bowel resection predominated. The proportion of emergency surgery and the mortality associated with emergency surgery was 30% higher in the oldest group (≥ 90) than in adults aged 18 to 64.

CONCLUSION

This article reports the pattern of surgical procedures, complications, and mortality found in NACOR, which is one of the few data sets that contains data from community hospitals and individuals with all types of insurance. Because the outcomes portion of the data set is under development, it is not possible to investigate the relationship between hospital type and complications or mortality, but this study underscores the magnitude of geriatric surgery that occurs in community hospitals as an area for future outcomes studies.

The effectiveness of specialist neuroscience care in severe traumatic brain injury: a systematic review

BACKGROUND

UK trauma services are currently undergoing reconfiguration, but the optimum management pathway for head-injured patients is uncertain. We therefore performed a systematic review to assess the effects of routine inter-hospital transfer and specialist neuroscience care on mortality and disability in patients with non-surgical severe traumatic brain injury injured nearest to a non-specialist acute hospital.

METHODS

A protocol was registered with PROSPERO (CRD42012002021) and review methodology followed Cochrane Collaboration recommendations. A peer reviewed search strategy was implemented in an exhaustive range of information sources, including all major bibliographic databases, between 1973 and July 2013. Selection of eligible studies, extraction of relevant data and bias assessment were then performed by two independent reviewers. In the absence of homogeneous effect estimates at low risk of bias a narrative synthesis was pre-specified.

RESULTS

Four cohort studies, including a total of 4688 patients, were identified as potentially eligible after screening and bias assessment. Confounding by indication, arising from selective transfer of less severely injured patients, was the main limitation of included studies, with overall risk of bias rated as high for both mortality and disability effect estimates. Adjusted odds ratios for mortality favoured secondary transfer, ranging from 1.92 (95% CI 1.25-2.95) to 2.09 (95% CI 1.59-2.74). No convincing association was observed between non-specialist care and unfavourable outcome with a conditional odds ratio of 1.13 (95% CI 0.36-3.6).

CONCLUSIONS

There is limited evidence supporting a strategy of secondary transfer of severe non-surgical traumatic brain injury patients to specialist neuroscience centres. Randomised controlled trials powered to detect clinically plausible treatment effects should be considered to definitively investigate effectiveness.

An integrated approach to evaluating alternative risk prediction strategies: a case study comparing alternative approaches for preventing invasive fungal disease

OBJECTIVES

This article proposes an integrated approach to the development, validation, and evaluation of new risk prediction models illustrated with the Fungal Infection Risk Evaluation study, which developed risk models to identify non-neutropenic, critically ill adult patients at high risk of invasive fungal disease (IFD).

METHODS

Our decision-analytical model compared alternative strategies for preventing IFD at up to three clinical decision time points (critical care admission, after 24 hours, and end of day 3), followed with antifungal prophylaxis for those judged "high" risk versus "no formal risk assessment." We developed prognostic models to predict the risk of IFD before critical care unit discharge, with data from 35,455 admissions to 70 UK adult, critical care units, and validated the models externally. The decision model was populated with positive predictive values and negative predictive values from the best-fitting risk models. We projected lifetime cost-effectiveness and expected value of partial perfect information for groups of parameters.

RESULTS

The risk prediction models performed well in internal and external validation. Risk assessment and prophylaxis at the end of day 3 was the most cost-effective strategy at the 2% and 1% risk threshold. Risk assessment at each time point was the most cost-effective strategy at a 0.5% risk threshold. Expected values of partial perfect information were high for positive predictive values or negative predictive values (£11 million-£13 million) and quality-adjusted life-years (£11 million).

CONCLUSIONS

It is cost-effective to formally assess the risk of IFD for non-neutropenic, critically ill adult patients. This integrated approach to developing and evaluating risk models is useful for informing clinical practice and future research investment.